Relationship between the attributes consists of nominal, polynominal and numerical data
<?xml version="1.0" encoding="UTF-8"?><process version="8.2.001">
<context>
<input/>
<output/>
<macros/>
</context>
<operator activated="true" class="process" compatibility="6.0.002" expanded="true" name="Process">
<process expanded="true">
<operator activated="true" class="retrieve" compatibility="8.2.001" expanded="true" height="68" name="Retrieve 8 Data Cleaning test 3" width="90" x="112" y="34">
<parameter key="repository_entry" value="8 Data Cleaning test 3"/>
</operator>
<operator activated="true" class="concurrency:correlation_matrix" compatibility="8.2.000" expanded="true" height="103" name="Correlation Matrix" width="90" x="447" y="238">
<parameter key="include_special_attributes" value="true"/>
</operator>
<connect from_op="Retrieve 8 Data Cleaning test 3" from_port="output" to_op="Correlation Matrix" to_port="example set"/>
<connect from_op="Correlation Matrix" from_port="example set" to_port="result 2"/>
<connect from_op="Correlation Matrix" from_port="matrix" to_port="result 1"/>
<portSpacing port="source_input 1" spacing="0"/>
<portSpacing port="sink_result 1" spacing="63"/>
<portSpacing port="sink_result 2" spacing="21"/>
<portSpacing port="sink_result 3" spacing="147"/>
</process>
</operator>
</process>
Best Answers
-
hello @azziaty256 - welcome to the community! Yes you can only use Correlation Matrix with numerical attributes as it is simply calculating pairwise r (or r^2 if you choose) correlation coefficients. It would be rather odd to calculate r (or r^2) between nominal / polynominal attributes as they are not numerical. I would recommend rethinking exactly what you are trying to do.
Scott
5 -
Hi,
In RapidMiner's Auto Model, there is some smart preprocessing done so that the correlations can also be calculated for your non-numerical attributes / columns. I would suggest to give it a try and see if this is what you are after. If yes, you can open the correlation process from Auto Model and adapt it to your needs.
Best,
Ingo
1
Answers
-
Sorry I am newbie..
Please help me how to make correlation between the attributes for a nominal, polynominal and numerical data types.
0 -
hello @azziaty256 - welcome to the community! Yes you can only use Correlation Matrix with numerical attributes as it is simply calculating pairwise r (or r^2 if you choose) correlation coefficients. It would be rather odd to calculate r (or r^2) between nominal / polynominal attributes as they are not numerical. I would recommend rethinking exactly what you are trying to do.
Scott
5 -
Hi,
In RapidMiner's Auto Model, there is some smart preprocessing done so that the correlations can also be calculated for your non-numerical attributes / columns. I would suggest to give it a try and see if this is what you are after. If yes, you can open the correlation process from Auto Model and adapt it to your needs.
Best,
Ingo
1 -
Thank you..ok i got it👍1